Protein folding, unfolding and misfolding are fundamental physical properties of biomaterials. As such, they play critical roles in defining protein stability, are the underlying cause of many disease states and serve as a potential source of both technological challenges and innovations. Despite the decades of scrutiny motivated by these important issues, however, no quantitative, experimentally verified hypothesis yet explains how protein folding occurs some 30+ orders of magnitude more rapidly than would a random conformational search. The goal of the proposed research program is to further our understanding of the folding process via the successful marriage of a novel, fully atomistic, massively parallel method for simulating protein folding with the experimental characterization of the relevant proteins under directly comparable conditions. Our motivation is straightforward. Experimental studies of the folding of simple proteins provide at best only a limited view of the folding transition-state. Simulations, in contrast, can provide an arbitrarily detailed representation of folding (limited only by the coupled issues of accuracy and computational complexity). They are, however, critically dependent on experimental validation and lack a clear, universally recognized means of identifying members of the transition state ensemble. By coupling simulation with experiment, we can circumvent these difficulties; the combination furnishes a method of "fleshing-out" the details otherwise hidden to experiment, provides an unimpeachable means of validating the simulations and offers guidance in the identification of the rate limiting transition. Inspired by these potential advantages, we propose here a novel, intimate coupling of simulation and experiment in an iterative process that significantly increases the utility of both.The historical difficulty with connecting simulation and experiment has been the conflict that small, rapidly folding proteins are difficult to examine experimentally, whereas large and/or slowly folding proteins are difficult to examine via detailed simulations. As significant preliminary results demonstrate, however, recent advances in both simulation and experiment now allow us to resolve this conflict. We have exploited these advances to reach and test multi-use, fully detailed folding simulations, providing a unique opportunity to directly compare the outcome of detailed simulations with experimental observations.